Table 10.

Advantages and disadvantages of swarm intelligence algorithms.

Algorithm	Advantages	Disadvantages
Particle Swarm Optimization (PSO)	good for continuous data high efficiency at finding a single global optimum fast convergence for simple search spaces user-friendly	struggles with discrete data lack of population diversity may become stuck in a local optimum poor performance with complex search spaces
Ant Colony Optimization (ACO)	good for handling discrete data good exploration and exploitation convenient for combinatorial search spaces robust to problem changes	struggles with large-scale problems slow convergence may become stuck in a local optimum sensitive to parameter choice
Whale Optimization Algorithm (WOA)	good for global search simple and user-friendly handles both continuous and discrete problems low number of control parameters converges quickly in continuous search spaces	may become stuck in local optima in multimodal search spaces sensitive to parameter setting struggles with exploration in irregular search spaces struggles with large-scale problems
Grey Wolf Optimizer (GWO)	applicable to both discrete and continuous problems good global search capabilities fast convergence simple parameter tuning overall simplicity and user-friendliness	struggles with large-scale problems may become stuck in a local optimum limited exploration capabilities in irregular search spaces sensitive to parameter tuning
Firefly Optimization Algorithm (FOA)	good global search capabilities good for both continuous and discrete problems fast convergence in continuous search spaces good for large-scale problems (parallelizable) simple and user-friendly	limited exploration capabilities in irregular search spaces may become stuck in a local optimum (premature convergence) poor population diversity sensitive to parameter tuning
Bat Optimization Algorithm (BOA)	good balance between exploration and exploitation good for both continuous and discrete problems includes local search capabilities fast convergence in continuous search spaces simple and user-friendly	limited exploration capabilities in irregular search spaces may become stuck in a local optimum struggles with large-scale optimization problems (poor scalability) sensitive to parameter tuning
Orca Predation Algorithm (OPA)	•good exploration/exploitation balance •ability to find multiple global solutions in multimodal problems good for problems with constraints	may struggle with global search may become stuck in a local optimum •poor diversification •may be prone to localization errors
Starling Murmuration Optimizer (SMO)	•good exploration/exploitation balance •good diversity •good for finding global optima, bypassing local minima •applicable to diverse search spaces •fast convergence	struggles with large-scale optimization problems (poor scalability) computational cost increases with the number of dimensions •relatively complex